Estimation of time-varying velocities of moving objects by time-frequency representations

An approach to the estimation of motion parameters of moving objects in a video-sequence, by using the SLIDE (subspace-based line detection algorithm) algorithm, is considered. The proposed procedure projects video-frames to the coordinate axes, in order to obtain synthetic images containing information about the motion parameters. These synthetic images are mapped to the FM signals by using constant /spl mu/-propagation. The problem of velocity estimation is reduced to the instantaneous frequency (IF) estimation. IF estimators, based on time-frequency (TF) representations, are used. Three TF representations: spectrogram (SPEC), Wigner distribution (WD), and S-method (SM), are used and compared to this aim. A tradeoff between concentration of the TF representation (velocity estimation accuracy) and reduction of the cross-terms (possibility for estimation of the multiple objects parameters) is achieved by the SM. A performance analysis of the algorithm is done. Theoretical results are illustrated on several numerical examples.     

基于改进的Camshift运动目标跟踪算法的研究

针对基于颜色概率分布的连续自适应均值漂移算法(Camshift)跟踪算法在背景中出现相同颜色干扰时容易致使跟踪目标失败的问题,提出了一种改进的Camshift跟踪算法。首先对Camshift跟踪目标前进行目标检测,通过帧差法、光流法、背景差分法三种检测算法对比,采用背景差分法得到的运动目标区域矩形特征参数作为Camshift的初始化参数,取代一般Camshift算法利用颜色特征的跟踪。最后对改进的算法和一般Camshift进行仿真对比实验。实验结果表明,结合背景差分法和连续Camshift算法的运动目标跟踪在一定程度上满足了实时性与稳定性的要求。     

运动目标的自动分割与跟踪

该文提出了一种对视频序列中的运动目标进行自动分割的算法。该算法分析图像在L U V空间中的局部变化,同时使用运动信息来把目标从背景中分离出来。首先根据图像的局部变化,使用基于图论的方法把图像分割成不同的区域。然后,通过度量合成的全局运动与估计的局部运动之间的偏差来检测出运动的区域,运动的区域通过基于区域的仿射运动模型来跟踪到下一帧。为了提高提取的目标的时空连续性,使用Hausdorff跟踪器对目标的二值模型进行跟踪。对一些典型的MPEG-4测试序列所进行的评估显示了该算法的优良性能。     

Iterative restoration of MR images corrupted with translational motion

Patient motion during the acquisition of magnetic resonance imaging data causes loss of resolution and ghost repetitions of the moving structures in the reconstructed image. In this paper the motion is modeled as being translational, and it is shown that this causes either the magnitude or the phase of the data to be corrupted, depending upon whether the motion is within or perpendicular to the imaging plane. The problem of restoring the image using only the corrupted data and no knowledge about the motion is addressed. The restoration problem is nonlinear in general, but is linear in two special cases. An iterative algorithm is developed that uses projections onto convex sets for magnitude retrieval and generalized projections for phase retrieval. In both cases constraint sets containing all a priori knowledge are used, and this is shown to be necessary for rapid convergence. The two algorithms may be combined to restore images corrupted by three-dimensional motion. The algorithms were verified using simulated data.     

复杂背景下多运动目标轮廓检测

该文在研究多种运动目标轮廓检测算法的基础上,提出了一种新的复杂背景下基于连续3帧即可精确检测多运动目标轮廓的算法。分析和实验表明,该算法抗干扰能力强,可以有效地消除被运动目标遮挡和重现的纹理背景,对复杂背景及不重叠多目标运动情况,可以精确地定位各个运动目标的外轮廓。此外,该算法具有潜在的并行机制,易于实现实时运动图像处理。     

Figure-ground segmentation from occlusion.

Layered video representations are increasingly popular; see [2] for a recent review. Segmentation of moving objects is a key step for automating such representations. Current motion segmentation methods either fail to segment moving objects in low-textured regions or are computationally very expensive. This paper presents a computationally simple algorithm that segments moving objects, even in low-texture/low-contrast scenes. Our method infers the moving object templates directly from the image intensity values, rather than computing the motion field as an intermediate step. Our model takes into account the rigidity of the moving object and the occlusion of the background by the moving object. We formulate the segmentation problem as the minimization of a penalized likelihood cost function and present an algorithm to estimate all the unknown parameters: the motions, the template of the moving object, and the intensity levels of the object and of the background pixels. The cost function combines a maximum likelihood estimation term with a term that penalizes large templates. The minimization algorithm performs two alternate steps for which we derive closed-form solutions. Relaxation improves the convergence even when low texture makes it very challenging to segment the moving object from the background. Experiments demonstrate the good performance of our method.     

Particle filter with a mode tracker for visual tracking across illumination changes

In this correspondence, our goal is to develop a visual tracking algorithm that is able to track moving objects in the presence of illumination variations in the scene and that is robust to occlusions. We treat the illumination and motion ( x-y translation and scale) parameters as the unknown "state" sequence. The observation is the entire image, and the observation model allows for occasional occlusions (modeled as outliers). The nonlinearity and multimodality of the observation model necessitate the use of a particle filter (PF). Due to the inclusion of illumination parameters, the state dimension increases, thus making regular PFs impractically expensive. We show that the recently proposed approach using a PF with a mode tracker can be used here since, even in most occlusion cases, the posterior of illumination conditioned on motion and the previous state is unimodal and quite narrow. The key idea is to importance sample on the motion states while approximating importance sampling by posterior mode tracking for estimating illumination. Experiments demonstrate the advantage of the proposed algorithm over existing PF-based approaches for various face and vehicle tracking. We are also able to detect illumination model changes, e.g., those due to transition from shadow to sunlight or vice versa by using the generalized expected log-likelihood statistics and successfully compensate for it without ever loosing track.     

3D Video Based Segmentation and Motion Estimation with Active Surface Evolution

In this paper, a novel method for three-dimensional (3D) segmentation and motion estimation based on 3D videos provided by TOF cameras is presented. The problem is formulated by a variational statement derived from the maximum a posterior probability (MAP) using 3D Optical Flow Constraint, containing both evolution surface and motion parameters. Therefore, the proposed method allows them to benefit from each other and perform motion segmentation and estimation simultaneously. All the formulation is under the assumption that environmental objects are rigid, and an iterative, PDE-driven level set method is adopted for energy minimization. Various experimental results show the validity of the proposed algorithm.     

Estimation of multiple 2-D uniform motions by SLIDE: subspace-basedline detection

A technique is proposed for estimating the parameters of two-dimensional (2-D) uniform motion of multiple moving objects in a scene, based on long-sequence image processing and the application of a multiline fitting algorithm. Plots of the vertical and horizontal projections versus frame number give new images in which uniformly moving objects are represented by skewed band regions, with the angles of the skew from the vertical being a measure of the velocities of the moving objects. For example, vertical bands will correspond to objects with zero velocity. An algorithm called subspace-based line detection (SLIDE) can be used to efficiently determine the skew angles. SLIDE exploits the temporal coherence between the contributions of each of the moving patterns in the frame projections to enhance and distinguish a signal subspace that is defined by the desired motion parameters. A similar procedure can be used to determine the vertical velocities. Some further steps must then be taken to properly associate the horizontal and vertical velocities.     

Time efficient ghost removal for motion detection in visual surveillance systems

Segmentation of foreground objects is an important and essential task for many systems that aim to carry out motion tracking, object classification, event detection and is used in applications such as traffic monitoring and analysis, access control to special areas, human and vehicle identification and the detection of anomalous behaviour. The most common approach for detecting moving objects is background subtraction, in which each frame of a video sequence is compared against a background model. A large number of background subtraction algorithms have been proposed [1], but problems remain for moving object identification under certain conditions. One of the toughest problems in background subtraction is caused by the detection of false objects when an object that belongs to the background (e.g. after staying stationary for some time) starts to move away. This generates what are called `ghosts?. It is important to address the problem because ghost objects will adversely affect many tasks such as object classification, tracking and event analysis (e.g. abandoned item detection). This Letter focuses on the problem of ghost identification and elimination. We used a state-of-the-art industrial tracker which includes basic background subtraction and object tracking. Then we included our ghost detection algorithm into the basic tracker to identify and eliminate ghosts. Finally, we systematically evaluated and compared performance on urban traffic video sequences.     

结合FCM聚类与SVM的火焰检测算法

针对传统视频型火焰检测算法误报率高、局限性强等问题,提出一种四步火焰检测算法。首先利用一种自适应混合高斯模型（GMM）检测视频序列中的运动目标;然后采用模糊C 均值（FCM）聚类算法分割疑似火焰区域与非火区域;再提取疑似火焰区域的面积变化、表面不均度等时空特征参数;最后将这些特征参数输入训练好的支持向量机（SVM）分类器以识别火焰区域。实验结果表明,算法不但在提高了检测率的同时降低了误检率,而且适用范围广,是一种有效的火焰检测算法。     

混合交通环境中的阴影检测算法

在城市交通流量视频检测系统中,目标阴影总是干扰对目标的正确检测和识别。在混合交通环境下,传统的阴影检测算法总是避免不了进行边缘检测、模板匹配等运算,不仅处理速度慢,而且对行人阴影的检测效果不好。本文提出一种基于颜色信息的阴影检测算法,该算法首先在图像中检测出运动区域,然后在运动区域内计算目标R、G、B颜色分量的灰度距离和色彩距离;最后根据这两个距离量检测出区域中的阴影。实验表明,该算法能够正确检测出车辆和行人的阴影,还能在雨天对目标的路面倒影进行检测,而且计算速度较快。     

空域视频场景监视中运动对象的实时检测与跟踪技术

本文分析了空域视频场景中运动对象实时检测、跟踪系统的模型。提出了一种在运动背景下实时检测与跟踪视频运动目标的技术。该方法首先进行背景的全局运动参数估计,并对背景进行补偿校正,将补偿校正后的相邻两帧进行差分检测。然后利用假设检验从差分图像中提取运动区域,利用遗传学方法在指定区域内确定最优分割门限,提取视频运动对象及其特征;最后利用线性预测器对目标进行匹配跟踪。在基于高速DSP的系统平台上的实验结果表明该方法取得了很好的效果。     

Unsupervised statistical neural networks for model-based objectrecognition

Statistical neural networks executing soft-decision algorithms have been shown to be very effective in many classification problems. A neural network architecture is developed here that can perform unsupervised joint segmentation and labeling of objects in images. We propose the semi-parametric hierarchical mixture density (HMD) model as a tool for capturing the diversity of real world images and pose the object recognition problem as a maximum likelihood (ML) estimation of the HMD parameters. We apply the expectation-maximization (EM) algorithm for this purpose and utilize ideas and techniques from statistical physics to cast the problem as the minimization of a free energy function. We then proceed to regularize the solution thus obtained by adding smoothing terms to the objective function. The resulting recursive scheme for estimating the posterior probabilities of an object's presence in an image corresponds to an unsupervised feedback neural network architecture. We present here the results of experiments involving recognition of traffic signs in natural scenes using this technique     

Automatic nonrigid calibration of image registration for real time MR-guided HIFU ablations of mobile organs

Real time magnetic resonance imaging (MRI) is rapidly gaining importance in interventional therapies. An accurate motion estimation is required for mobile targets and can be conveniently addressed using an image registration algorithm. Since the adaptation of the control parameters of the algorithm depends on the application (targeted organ, location of the tumor, slice orientation, etc.), typically an individual calibration is required. However, the assessment of the estimated motion accuracy is difficult since the real target motion is unknown. In this paper, existing criteria based only on anatomical image similarity are demonstrated to be inadequate. A new criterion is introduced, which is based on the local magnetic field distribution. The proposed criterion was used to assess, during a preparative calibration step, the optimal configuration of an image registration algorithm derived from the Horn and Schunck method. The accuracy of the proposed method was evaluated in a moving phantom experiment, which allows the comparison with the known motion pattern and to an established criterion based on anatomical images. The usefulness of the method for the calibration of optical-flow based algorithms was also demonstrated in vivo under conditions similar to thermo-ablation for the abdomen of twelve volunteers. In average over all volunteers, a resulting displacement error of 1.5 mm was obtained (largest observed error equal to 4-5 mm) using a criterion based on anatomical image similarity. A better average accuracy of 1 mm was achieved using the proposed criterion (largest observed error equal to 2 mm). In both kidney and liver, the proposed criterion was shown to provide motion field accuracy in the range of the best achievable.     

Motion Estimation Algorithms Using the Deformation of Planar Hierarchical Mesh Grid for Videoconferencing Applications at Low Bit-rate Transmission

This paper studies the issue of reducing the temporal redundancy between consecutive frames of a videoconferencing sequence at low bit-rate transmission. To overcome the drawbacks of the traditional block matching algorithm implemented in the most current video coding standards, we propose to better describe the motion of objects through the deformation of planar rectangular mesh grid adapted to the edges of the moving objects in the scene. The traditional inter coding modes are then replaced by two new classes of encoding algorithms. The first one concerns the B-frames where the problem of motion estimation is solved by a bidirectional prediction algorithm which reconstructs the quadrilateral mesh grids without any coding cost. The second class of algorithm much more complex than the first one is specific to the P-frames based on the principle of merging two hierarchical grids of reference. This algorithm addresses not only the motion estimation problem based on the adaptive quadrilateral mesh grid but also the issue of the relevant information (e.g. the positions of the nodes, the connectivity of each quadrilateral mesh of the grid and the motion compensation) to efficiently encode. The implementation of these algorithms in a complete coding scheme offers good performance compared to the H.264/AVC video coder at low bit-rate transmission.     

针对运动摄像机的快速低存储开销运动目标检测算法

摄像机的运动会导致整幅图像的运动,使得此情形下的目标检测极具挑战性。针对该问题提出一种快速低存储开销检测算法。首先,利用一种快速低存储开销配准方法计算相邻两帧的单应变换矩阵。而后,使用单应变换矩阵进行相邻两帧之间的配准,并由帧间差分获取帧间运动信息。最后,采用积累运动信息的方式构造不断更新的运动图像,通过对此运动图像进行阈值分割分离出最终的运动目标。在多个不同视频序列下的实验表明该算法能够有效地从嘈杂的场景中检测出运动目标。此外,与先前算法相比,该算法检测性能更好,且显著地降低了存储开销与计算时间开销。对于480360的序列而言,该算法需要的存储开销仅为825 kByte,且运算速度达到16帧/m。     

基于运动目标的De-Interlace算法

本文提出一种新颖的基于运动目标的De-interlace算法.该算法以实际的运动目标作为操作对象,引入一种较精确的运动目标提取方法,并采用免疫克隆选择算法加速匹配目标的搜索过程.新算法融合了运动补偿、中值滤波、Weave、Bob等De-interlace方法.与流行的基于运动块补偿的De-interlace算法相比,新算法更适应复杂的视频序列,不仅可以处理平移运动,还适用于旋转、尺度变换等复杂运动情况.实验结果表明新算法的整体性能优于基于块匹配的方法.